Is a Macroprudential Reaction Function Emerging and Is it Sensible?

A decade on from the global financial crisis, is a macroprudential reaction function emerging and is it sensible? We address this question in two parts. First, we assess the responsiveness of the countercyclical capital buffer (CCyB) and housing tools to developments in the risk environment. We find that both tools are tightened in response to rapid household credit growth, with a one standard deviation increase approximately doubling the probability that tools are tightened in the following year. CCyB use also responds to rapid house price growth and low equity volatility. Second, we compare the weight policymakers have given to different risk indicators in practice with the “GDP-at-risk” weight of each indicator, based on its predictive power over severe recessions in the past. We find that GDP-at-risk weights produce a useful aggregate indicator with which to predict tools use. However, in practice, countries have tended to overweight household credit relative to its GDP-at-risk weight and put too little weight on other indicators, including households’ debt service ratios and corporate credit growth. While the risk indicators we consider can explain around 30% of variation in the CCyB, their explanatory power over housing tools is much more limited. This may indicate that housing tools are in part put in place as structural guardrails rather than to address current vulnerabilities.